Development and Application of Physiologically-Based Pharmacokinetic Model to Predict Systemic and Organ Exposure of Colorectal Cancer Drugs.

BACKGROUND/OBJECTIVES: Colorectal cancer (CRC) holds the third and second position among cancers affecting men and women, respectively. Frequently, the first-line treatment for metastatic CRC consists of the intravenous administration of 5-fluorouracil and leucovorin in combination with oxaliplatin or irinotecan. Physiologically-based pharmacokinetic models (PBPK) aim to mechanistically incorporate body physiology and drug physicochemical attributes, enabling the description of both systemic and organ drug exposure based on the treatment specificities. This bottom-up approach represents an opportunity to personalize treatment and minimize the therapeutic risk/benefit ratio through the understanding of drug distribution within colorectal tissue. This project has the goal of characterizing the systemic and tissue exposure of four anti-cancer drugs in humans using a PBPK platform fed with data from the literature.

METHODS

A literature search was performed to collect clinical data on systemic concentration versus time profiles. Physicochemical features were obtained from the literature, as well as parameters associated with distribution, metabolism, and excretion. The PBPK models were built using PK-Sim.

RESULTS

The data from 51 clinical studies were extracted and combined in one single dataset. The PBPK models successfully described the exposure vs. time profiles with respect to both, with both the typical tendency and dispersion shown by the data. The percentage of observations falling within the two-fold error bounds ranged between 94 and 100%. The colon/plasma AUC ratios were similar for 5-FU, oxaliplatin, and leucovorin, but it was significantly higher for irinotecan.

CONCLUSIONS

The PBPK models support tailored treatment approaches by linking in vitro studies to organ exposure. These models serve as the initial step towards incorporating a dedicated tumor compartment, which will further account for the variability in tumor microenvironment characteristics to improve therapeutic strategies.